Hypothesis '(H) theory characterized human problem solving (generally restricted to the discrimination-learning and concept-learning tasks) as follows: the S learns by testing Hs according to aamore-or-lessefficient strategy. Probes for detecting S's H are well developed; techaniques for detecting his strategy (his system for selecting Hs to test) are being refined. The general objective of the proposed program is to expand the theory both by incorporating processes other than H testing which contribute to problem solving and by applying the theory to a variety of problem-solving situations. The immediate steps to be taken (using adults and children of different ages) are: (1) The study of transfer over a series of learning or problem-solving tasks. (2) The analysis of the detailed component processes (e.g., imagery and other analysis of the detailed component processes (e.g., imagery and other mnemonic technigues) underlying the H-testing process. (3) The intensive study of individual differences. This will serve to validate the relationship between the component processes and problem-solving ability. (4) The determination of methods (instructions, motivation variables) for improving the strategies, particularly those of young Ss.